Macromolecules have demonstrated great value as research tools, human therapeutics, and diagnostics. The size and complexity of folded macromolecules can result in potencies and specificities of action that are not easily achievable using small molecules. Features unique to folded macromolecules make them well suited for therapeutic and basic research applications. The theoretical functional diversity of human health-related antibodies, or their fragments, is large because antibodies and antibody fragments can be raised against a theoretically infinite number of disease-related targets in vivo, including targets that are typically viewed as undruggable using small molecules. However, the generation, production, and purification of antibodies and their fragments is often difficult and expensive. Methods to identify unnatural proteins that bind therapeutically relevant cellular targets in vivo are of considerable interest to the biomedical community because they may significantly increase the number of proteins with therapeutic activity, and represent an alternative strategy to antibody-based macromolecular therapeutic development. Here, we propose to exploit our recent development of split-superpositive GFP reassembly to identify unnatural proteins with affinity for ankyrin repeat domain proteins overexpressed in disease, and which frustrate small molecule-based drug discovery. In addition, we will use protein evolution and engineering to identify RNA binding proteins with high affinity and specificity for RNA hairpins critical to HIV replication. The therapeutic potential of these new protein reagents will be measured using a number of in vitro and cell-based assays. Taken together, the findings revealed in this research will potentially lead to the generation of proteins that overcome limitations to small molecule-centered drug discovery, and may point the way to new therapeutic strategies.